Frequency-diversity technique for reliable radar level measurement of bulk solids in silos

In this paper a concept for multi-spectral analysis of radar echo signals is presented, which is aimed at level measurement of bulk goods in silos. Severe problems given in these applications result from the conditions that bulk goods build up a heap and that the microwaves are backscattered at both, its complicated surface and also from inside the heap. The proposed frequency-diversity technique has been evaluated by means of echo measurements in silos filled with plastic pellets and chalk powder. A Frequency Modulated Continuous Wave (FMCW) radar system with 25 GHz center frequency and 2 GHz bandwidth has been used for the measurements. Results show that echo signal fluctuations, which are very strong with the conventional single-band approach, are now largely minimized, what allows for robust and reliable level measurement. Furthermore, we suggest a variance analysis of echo signals from the multiple frequency sub-bands as a method for classification. It will also be shown with experimental results that this strategy can successfully be applied to distinguish between echoes caused by backscattering from randomly arranged inhomogeneities in the bulk solids and spatially separated disturbing radar-targets. The latter are for example given by weld seams at the silo wall, whose echoes may cause severe disturbances of the level measurement.